skills-reference
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: Uses dynamic context injection via the !command syntax to execute shell commands like git log, gh pr diff, and git branch at skill load time to gather project information.
- [COMMAND_EXECUTION]: Includes instructions for the agent to execute local shell scripts (e.g., bash scripts/check-prerequisites.sh) and run environment setup commands like npm install and npm run db:setup.
- [EXTERNAL_DOWNLOADS]: Invokes npx prisma validate, which can trigger the download of the prisma package and its dependencies from the official npm registry.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes external content. Ingestion points: Untrusted data enters the agent context through $ARGUMENTS and the output of the gh pr diff and gh pr view commands. Boundary markers: The templates do not use explicit delimiters or instructions to ignore embedded instructions within the ingested PR or commit data. Capability inventory: The skill uses Bash, Write, and Read tools across its example scripts. Sanitization: No validation or sanitization is performed on the external data before it is processed or interpolated into prompts.
Audit Metadata